Acidic and Basic Side Chains

The side chains of aspartic and glutamic acids carry negatively charged carboxylate groups at pH 7 while those of lysine and arginine carry the positively charged –NH3 + and guanidinium ions, respectively.

At pH 7 the weakly basic imidazole group of histidine may be partially protonated. Both the –SH group of cysteine and the phenolic –OH of tyrosine are weakly acidic and will dissociate and thereby acquire negative charges at a sufficiently high pH.
The number of positive and negative charges on a protein at any pH can be estimated approximately from the acid dissociation constants (usually given as pKa values) for the amino acid side chains. However, pKa values of buried groups are often greatly shifted from these, especially if they associate as ion pairs. In addition, many proteins have free amino and carboxyl-terminal groups at the opposite ends of the peptide chain. These also participate in acid–base reactions with approximately the following pKa values.

terminal --, pKa = 3.6–3.7
terminal --NH3+, pKa = 7.5–7.9

The acid–base properties of an amino acid or of a protein are described by titration curves of the type. In these curves the number of equivalents of acid or base that have reacted with an amino acid or protein that was initially at neutral pH are plotted against pH. The net negative or positive electrical charge on the molecule can be read directly from the curves. Both the net electrical charges and the distribution of positively and negatively charged groups are often of crucial importance to the functioning of a protein.

Properties of -Amino Acids

The amino acids have in common a dipolar ionic structure and a chiral center. They are differentiated, one from another, by the structures of their side chain groups, designated R in the foregoing formulas. These groups are of varying size and chemical structure. The side chain groups fill much of the space in the interior of a protein molecule and also protrude from the external surfaces of the protein where they determine many of the chemical and physical properties of the molecule.
Show the structures of the side chains of the amino acids commonly found in proteins. The
complete structure is given for proline. Both the threeletter abbreviations and one-letter abbreviations used in describing sequences of amino acids in proteins are also given in this table. Amino acids of groups a–c plus phenylalanine and methionine are sometimes grouped together as nonpolar. They tend to be found in a hydrophobic environment on the inside of a protein molecule. Groups f and i contain polar, charged side chains which usually protrude into the water surrounding the protein. The rest are classified as polar but noncharged.
To get acquainted with amino acid structures, learn first those of glycine, alanine, serine, aspartic acid, and glutamic acid. The structures of many other amino acids can be related to that of alanine (R=CH3) by replacement of a β hydrogen by another group. Metabolic interrelationships will make it easier to learn structures of the rest of the amino acids later.

Since the –COOH groups of glutamic and aspartic acids are completely dissociated to –COO– at neutral pH, it is customary in the biochemical literature to refer to these amino acids as glutamate and aspartate without reference to the nature of the cation or cations present as counter ions. Such “-ate” endings are also used for most other acids (e.g., malate, oxaloacetate, phosphate, and adenylate) and in names of enzymes (e.g., lactate
dehydrogenase).
During the formation of polypeptides, the α-amino and carboxyl groups of the amino acids are converted into the relatively unreactive and uncharged amide (peptide)
groups except at the two chain termini. In many cases the terminal amino and carboxyl groups are also converted within cells into uncharged groups (Chapter 10). Immediately
after the protein is synthesized its terminal carboxyl group is often converted into an amide. The N terminus may be acetylated or cyclized to a pyroglutamyl group. Sometimes a cyclic peptide is formed.
The properties of polypeptides and proteins are determined to a large extent by the chemistry of the side chain groups, which may be summarized briefly as follows. Glycine in a peptide permits a maximum of conformational mobility. The nine relatively nonpolar amino acids–alanine, valine, leucine, isoleucine, proline, methionine, phenylalanine, tyrosine, and tryptophan– serve as building blocks of characteristic shape. Tyrosine and tryptophan also participate in hydrogen bonding and in aromatic: aromatic interactions within proteins.
Much of the chemistry of proteins involves the side chain functional groups –OH, –SH, –COO–, –NH3 +, and imidazole and the guanidinium group of arginine. The side chains of asparagine and glutamine both contain the amide group CONH2, which is relatively inert chemically but which can undergo hydrogen-bonding interactions. The amide linkages of the polypeptide backbone must also be regarded as important functional groups. Most polar groups are found on the outside surfaces of proteins where they can react chemically in various ways. When inside proteins they form H-bonds to the peptide backbone and to other polar groups.

Amino Acids and Peptides

Twenty α-amino acids are the monomers from which proteins are made. These amino acids share with other biochemical monomers a property essential to their role in polymer formation: They contain at least two different chemical groups able to react with each other to form a covalent linkage. In the amino acids these are the protonated amino (NH3 +) and carboxylate (COO–) groups. The characteristic linkage in the protein polymer is the peptide (amide) linkage whose formation can be imagined to occur by the splitting out of water between the carboxyl of one amino acid and the amino group of another.
This equation is not intended to imply a mechanism for peptide synthesis. The equilibrium position for this reaction in an aqueous solution favors the free amino acids rather than the peptide. Therefore, both biological and laboratory syntheses of peptides usually do not involve a simple splitting out of water. Since the dipeptide still contains reactive carboxyl and amino groups, other amino acid units can be joined by additional peptide linkages to form polypeptides. These range from short-chain oligomers to polymers of from ∼50 to several thousand amino acid units, the proteins.

Hydrophobic Interactions

Fats, hydrocarbons, and other materials whose molecules consist largely of nonpolar groups have a low solubility in water and a high solubility in nonpolar solvents. Similarly, the long alkyl groups of fatty acid esters aggregate within membranes and nonpolar side chains of proteins are often packed together in the centers of protein molecules. Because it is as if the nonpolar groups “fear” water, this is known as the hydrophobic effect. The terms hydrophobic forces, hydrophobic interactions, and hydrophobic bonding have also been used. However, the latter term can be misleading because the hydrophobic effect arises not out of any special attraction between nonpolar groups but primarily from the strong internal cohesion of the hydrogen-bonded water structure.

–CH2–(water) --→–CH2–(hexane)
ΔG° = –3.8 kJ/mol

This equation is a quantitative statement of the fact that the CH2 group prefers to be in a nonpolar environment than to be surrounded by water. A similar Gibbs energy change would be expected to accompany the bringing together of a methylene unit from a small molecule and a hydrophobic surface on a protein molecule. However, in the latter case the accompanying What causes the decrease in Gibbs energy when nonpolar groups associate in water? Jencks60 suggested that we think of the transfer of a nonpolar molecule from a nonpolar solvent into water in two steps: (1) Create a cavity in the water of about the right size to accommodate the molecule. Since many hydrogen bonds will be broken, the Gibbs energy of cavity formation will be high. It will be principally an enthalpy (ΔH) effect. (2) Allow the water molecules in the solvent to make changes in their orientations to accommodate the nonpolar molecule that has been placed in the cavity. The water molecules can move to give good van der Waals contacts and also reorient themselves to give the maximum number of hydrogen bonds. Since hydrogen bonds can be formed in many different ways in water, there may be as many or even more hydrogen bonds after the reorientation than before. This will be true especially at low temperature where most water exists as large icelike clusters. For dissolved hydrocarbons, the enthalpy of formation of the new hydrogen bonds often almost exactly balances the enthalpy of creation of the cavity initially so that ΔH for the overall process (transfer from inert solvent into water) is small. For the opposite transfer ΔH° is usually a small positive number for aliphatic hydrocarbons and is nearly zero for aromatic hydrocarbons.
Since ΔG° = ΔH°– TΔS°, it follows that the negative value of ΔG° for hydrophobic interactions must result from a positive entropy change, which may arise from the restricted mobility of water molecules that surround dissolved hydrophobic groups. When two hydrophobic groups come together to form a “hydrophobic bond,” water molecules are freed from the structured region around the hydrophobic surfaces and the entropy increases. The ΔS°. Attempts have been made to relate this value directly to the increased number of orientations possible for a water molecule when it is freed from the structured region. However, interpretation of the hydrophobic effect is complex and controversial.
The formation constant Kf for hydrophobic associations often increases with increasing temperature. This is in contrast to the behavior of Kf for many association reactions that involve polar molecules and for which ΔH° is often strongly negative (heat is released). An example of the latter is the protonation of ammonia in an aqueous solution

NH3 + H+ → NH4+
ΔH° = –52.5 kJ/mol

Since R1nKf = –ΔG°/T = –ΔH°/T + ΔS°, Kf decreases with increasing temperature if ΔH° is negative. Because for a hydrophobic interaction with a positive value of ΔH° Kf increases with increasing temperature, an increase in stability at higher temperatures is sometimes used as a criterion for hydrophobic bonding. However, this criterion does not always hold. For example, base stacking interactions in polynucleotides, whose strength does not increase with increasing temperature, are still thought to be hydrophobic.
The water molecules that are in immediate contact with dissolved nonpolar groups are partially oriented. They form a cagelike structure around each hydrophobic group. When particles surrounded by such hydration layers are 1–2 nm apart, they sometimes experience either a fairly strong repulsion or an enhanced attraction caused by these hydration layers Direct experimental measurements have shown that these effects extend to distances of 10 nm and can account for the previously mentioned long-range van der Waals forces.
Various efforts have been made to develop scales of hydrophobicity that can be used to predict the probability of finding a given amino acid side chain buried within a protein or in a surface facing water. A new approach has been provided by the study of mutant proteins. For example, deletion of a single –CH2– group from an interior hydrophobic region of a protein was observed to decrease the stability of the protein by 4.6 kJ/mol.

Hydration of Polar Molecules and Ions

Water molecules are able to hydrogen bond not only to each other but also to polar groups of dissolved compounds. Thus, every group that is capable of forming a hydrogen bond to another organic group is also able to form hydrogen bonds of a somewhat similar strength with water. For this reason, hydrogen bonding is usually not a significant force in holding small molecules together in aqueous solutions. Polar molecules that stick together through hydrogen bonding when dissolved in a nonpolar solvent often do not associate in water. How then can biochemists assert that hydrogen bonding is so important in biochemistry? Part of the answer is that proteins and nucleic acids can be either properly folded with hydrogen bonds
formed internally or denatured with hydrogen bonds from those same groups to water. The Gibbs energy change between these two states is small.
Every ion in an aqueous solution is surrounded by a shell of oriented water molecules held by the attraction of the water dipoles to the charged ion. The hydration of ions has a strong influence on all aspects of electrostatic interactions and plays a dominant role in determining such matters as the strength of acids and bases, the Gibbs energy of hydrolysis of ATP, and the strength of bonding of metal ions to negatively charged groups. For example, the previously considered interaction between carboxylate and calcium ions would be much weaker if both ions retained their hydration shells.
Consider the following example. ΔG° for dissociation of acetic acid in water is +27.2 kJ/mol. The enthalpy change ΔH° for this process is almost zero (–0.1 kJ/mol) and ΔS° is consequently –91.6 J K–1. This large entropy decrease reflects the increased amount of water that is immobilized in the hydration spheres of the H+ and acetate– ions formed in the dissociation reaction. In contrast, dissociation of NH4 + to NH3 and H+ converts one positive ion to another. ΔH° is large (+52.5 kJ/mol) but the entropy change ΔS° is small.
Although effects of hydration are important in almost all biochemical equilibria, they are difficult to assess quantitatively. It is hard to know how many molecules of water are freed or immobilized in a given reaction. Charged groups in proteins are often hydrated. However, if they are buried in the interior of the protein, they may be solvated by polarizable protein side chain groups such as –OH or by backbone or side chain amide groups.

The Structure and Properties of Water

Water is the major constituent of cells and a remarkable solvent whose chemical and physical properties affect almost every aspect of life. Many of these properties are a direct reflection of the fact that most water molecules are in contact with their neighbors entirely through hydrogen bonds.Water is the only known substance for which this is true.
In ordinary ice all of the water molecules are connected by hydrogen bonds, six molecules forming a hexagonal ring resembling that of cyclohexane. The structure is extended in all directions by the formation of additional hydrogen bonds to adjacent molecules. As can be seen in this drawing, the molecules in ice assume various orientations in the hexagonal array, and frequently rotate to form their hydrogen bonds in different ways. This randomness remains as the temperature is lowered, and ice is one of few substances with a residual entropy at absolute zero. Ice is unusual also in that the molecules do not assume closest packing in the crystal but form an open structure. The hole through the middle of the hexagon and on through the hexagons lying below it is ∼0.06 nm in diameter.
The short hydrogen-bond length (averaging 0.276 nm) in ice indicates of strong bonding. The heat of sublimation (ΔH°) of ice is –48.6 kJ/mol. If the van der Waals dispersion energy of –15 kJ/mol is subtracted from this, the difference of –34 kJ/mol can be attributed entirely to the hydrogen bonds—two for each molecule. Their average energy is 17 kJ/mol apiece. However, some of the hydrogen bonds are stronger and others weaker than the average.

Six water molecules in the lattice of an ice crystal. The hydrogen bonds, which connect protons with electron pairs of adjacent molecules, are shown as dashed lines.

In a gaseous water dimer the hydrogen bond is linear, a fact that suggests some covalent character.

Its length is distinctly greater than that in ice. This is one of a number of pieces of evidence suggesting cooperativity in formation of chains of hydrogen bonds. Consider the following three trimers for which theoretical calculations have predicted the indicated hydrogen bond energies. In the first case the central water molecule donates two protons for hydrogen-bond formation; in the second it accepts the protons. In the third case it is both an electron acceptor and a donor. The OH dipoles are oriented “head to tail” and the hydrogen bonds are stronger than in the other



cases. Long chains of similarly oriented hydrogen bonds exist in ice and this may account for the short hydrogen bond lengths. Closed rings of hydrogen bonds oriented to give a maximum cooperative effect also exist in liquid water clusters and within proteins, carbohydrates, and nucleic acids.
The nature of liquid water is still incompletely understood,but we know that water contains icelike clusters of molecules that are continually breaking up and reforming in what has been called a “flickering cluster” structure. Judging by the infrared spectrum of water, about 10% of the hydrogen bonds are broken when ice melts.41 A similar conclusion can be drawn from the fact that the heat of melting of ice is –5.9 kJ/mol. It has been estimated that at 0°C the average cluster contains about 500 water molecules.41 At 50°C there are over 100 and at the boiling point about 40. Although most molecules in liquid water are present in these clusters, the hydrogen bonds are rapidly broken and reformed in new ways, with the average lifetime of a given hydrogen bond being ∼10(–12) s.

Hydrogen Bonds

One of the most important weak interactions between biologically important molecules is the hydrogen bond (H-bond). These “bonds” are the result of electrostatic attraction caused by the uneven distribution of electrons within covalent bonds. For example, the bonding electron pairs of the H–O bonds of water molecules are attracted more tightly to the oxygen atoms than to the hydrogen atoms. A small net positive charge is left on the hydrogen and a small net negative charge on the oxygen. Such polarization of the water molecules can be indicated in the following way:



Here the δ+ and δ– indicate a fraction of a full charge present on the hydrogen atoms and on the nonbonded electron pairs of the oxygen atom, respectively. Molecules such as H2O, with strongly polarized bonds, are referred to as polar molecules and functional groups with such bonds as polar groups. They are to be contrasted with such nonpolar groups as the –CH3 group in which the electrons in the bonds are nearly equally shared by carbon and hydrogen.
A hydrogen bond is formed when the positively charged end of one of the dipoles (polarized bonds) is attracted to the negative end of another dipole. Water molecules tend to hydrogen bond strongly one to another; each oxygen atom can be hydrogen-bonded to two other molecules and each hydrogen to another water molecule. Thus, every water molecule can have up to four hydrogen-bonded neighbors.

A water molecule hydrogen bonded to four other water molecules; note the tetrahedral arrangement of bonds around the central oxygen.

Many groups in proteins, carbohydrates, and nucleic acids form hydrogen bonds to one another and to surrounding water molecules. For example, an imidazole group of a protein can bond to an OH group of an amino acid side chain or of water in the following
ways:



Remember that hydrogen bonds are always formed between pairs of groups, with one of them, often C=O or C=N-, containing the negative end of a dipole and the other providing the proton. The proton acceptor group, often OH or NH and occasionally SH, and even CH in certain structures,donates an unshared pair of electrons. Dashed arrows are sometimes drawn from the hydrogen atom to the electron donor atom to indicate the direction of a hydrogen bond. Do not confuse these arrows with the curved arrows that indicate flow of electrons in organic reactions.
The strength of hydrogen bonds, as measured by the bond energy, varies over the range 10–40 kJ/mol. The stronger the hydrogen bond the shorter its length. Because hydrogen atoms can usually not be seen in X-ray structures of macromolecules, the lengths of
hydrogen bonds are often measured between the surrounding heavy atoms:



A typical —OH- - -O hydrogen bond will have a length of about 0.31 nm; a very strong hydrogen bond may be less than 0.28 nm in length, while weak hydrogen bonds will approach 0.36 nm, which is the sum of the van der Waals contact distances plus the O–H bond length. Beyond this distance a hydrogen bond cannot be distinguished easily from a van der Waals contact.
Hydrogen bonds are strongest when the hydrogen atom and the two heavy atoms to which it is bonded are in a straight line. For this reason hydrogen bonds tend to be linear. However, the dipoles forming the hydrogen bond do not have to be colinear for strong hydrogen bonding: There is some preference for hydrogen bonding to occur in the direction of an unshared electron pair on the oxygen or nitrogen atom.

A linear O–H- - -O hydrogen bond with dipoles at an angle one to another.

Both ammonia, NH3, and the –NH2 groups of proteins are good electron donors for hydrogen bond formation. However, the hydrogen atoms of uncharged –NH2 groups tend to be poor proton donors for H-bonds. Do hydrogen bonds have some covalent character? The answer is controversial.
Hydrogen bonding is important both to the internal structure of biological macromolecules and in interactions between molecules. Hydrogen bonding often provides the specificity necessary to bring surfaces together in a complementary way. Thus, the
location of hydrogen-bond forming groups in surfaces between molecules is important in ensuring an exact alignment of the surfaces.37 The hydrogen bonds do not always have to be strong. For example, Fersht and coworkers, who compared a variety of mutants of an enzyme of known three-dimensional structure, found that deletion of a side chain that formed a good hydrogen bond to the substrate weakened the binding energy by only 2–6 kJ/mol. However, loss of a hydrogen bond to a charged group in the substrate caused a loss of 15–20 kJ/mol of binding energy. Study of mutant proteins created by genetic engineering is now an important tool for experimentally investigating the biological roles of hydrogen bonding.